suse-edge/Factory
SHA256
17
0
Fork 17
Code Issues Pull Requests 3 Actions Packages Projects Releases Wiki Activity

Update metal3 components to latest #317

Merged
nbelouin merged 3 commits from nbelouin/Factory:metal3-update into main 2025-12-12 08:19:46 +01:00
Conversation 6 Commits 3 Files Changed 35 +267 -918
nbelouin commented 2025-11-28 15:49:57 +01:00
Owner
Copy Link
No description provided.
nbelouin added 3 commits 2025-11-28 15:49:57 +01:00
Update BMO to 0.11.2 0b713dcd4b 
Signed-off-by: Nicolas Belouin <nicolas.belouin@suse.com>
Update to ironic-image v32 c0c1c662ce 
Signed-off-by: Nicolas Belouin <nicolas.belouin@suse.com>
Update metal3 chart
All checks were successful
Check Release Manifest Local Charts Versions / Check Release Manifest Local Charts Versions (pull_request) Successful in 9s
Details
Build PR in OBS / Build PR in OBS (pull_request_target) Successful in 31m8s
Details
ab71a0c3d7 
Signed-off-by: Nicolas Belouin <nicolas.belouin@suse.com>
steven.hardy reviewed 2025-12-05 15:43:09 +01:00
baremetal-operator/baremetal-operator.spec Outdated
@@ -18,3 +18,3 @@
Name: baremetal-operator
Version: 0.10.2
Version: 0.11.2
steven.hardy commented 2025-12-05 15:43:09 +01:00
Owner
Copy Link

Rancher 2.13 plans to stick with CAPI 0.10.x AFAIK because there are some disruptive changes in 0.11.x - so I'm wondering it it may be safer to stick with 0.10.x here? There is a newer 0.10.5 we could update to?

I know CAPM3 has coupling to core CAPI versions and documents a compatibility matrix so I've stuck with CAPM3 0.10.4 in rancher-turtles-providers, I don't think the coupling with BMO is as significant but perhaps it's safer to keep everything aligned on 0.10.x?

Rancher 2.13 plans to stick with CAPI 0.10.x AFAIK because there are some disruptive changes in 0.11.x - so I'm wondering it it may be safer to stick with 0.10.x here? There is a newer 0.10.5 we could update to? I know CAPM3 has coupling to core CAPI versions and documents a [compatibility matrix](https://github.com/metal3-io/cluster-api-provider-metal3/?tab=readme-ov-file#compatibility-with-cluster-api) so I've stuck with CAPM3 0.10.4 in rancher-turtles-providers, I don't think the coupling with BMO is as significant but perhaps it's safer to keep everything aligned on 0.10.x?
nbelouin commented 2025-12-08 09:01:35 +01:00
Author
Owner
Copy Link

AFAIK, there is no real coupling between BMO and CAPI versions here, apart from tests. I'm a bit reluctant to update to 0.10.5 as we were already pulling a few patches to backport features from 0.11.x

We should stick to CAPM3 0.10.4 however, as we are more likely to have issues there.

I'm a bit surprised with us sticking to CAPI 0.10.x though, as I thought only 0.11.x was supported for kubernetes 1.34

AFAIK, there is no real coupling between BMO and CAPI versions here, apart from tests. I'm a bit reluctant to update to 0.10.5 as we were already pulling a few patches to backport features from 0.11.x We should stick to CAPM3 0.10.4 however, as we are more likely to have issues there. I'm a bit surprised with us sticking to CAPI 0.10.x though, as I thought only 0.11.x was supported for kubernetes 1.34
steven.hardy marked this conversation as resolved
steven.hardy reviewed 2025-12-05 15:46:58 +01:00
steven.hardy left a comment
Owner
Copy Link

Overall looks good but I have a question about the target BMO version

Also have all the ironic-config scripts been rebased by copying upstream?

IIRC @mchiappero made some changes downstream which have not yet been accepted upstream, are we losing those as a result of this rebase? It would be easier if we applied such downstream changes as a patch in future perhaps?

Overall looks good but I have a question about the target BMO version Also have all the ironic-config scripts been rebased by copying upstream? IIRC @mchiappero made some changes downstream which have not yet been accepted upstream, are we losing those as a result of this rebase? It would be easier if we applied such downstream changes as a patch in future perhaps?
steven.hardy requested review from mchiappero 2025-12-05 15:47:09 +01:00
nbelouin commented 2025-12-08 09:04:11 +01:00
Author
Owner
Copy Link

@steven.hardy I updated ironic-image by doing a "manual rebase", a.k.a I applied all commits from previous version to actual version to our tree, I didn't copy the scripts as is. So all IPv6 related changes should still be there and working.

@steven.hardy I updated ironic-image by doing a "manual rebase", a.k.a I applied all commits from previous version to actual version to our tree, I didn't copy the scripts as is. So all IPv6 related changes should still be there and working.
👍 1
steven.hardy requested review from steven.hardy 2025-12-08 11:31:04 +01:00
steven.hardy approved these changes 2025-12-08 11:31:11 +01:00
mchiappero commented 2025-12-11 14:32:44 +01:00
Owner
Copy Link

At some point I will have to bring back the changes from upstream, where I have improved the code quite a bit.
That said, it LGTM.

At some point I will have to bring back the changes from upstream, where I have improved the code quite a bit. That said, it LGTM.
mchiappero approved these changes 2025-12-11 14:43:31 +01:00
nbelouin force-pushed metal3-update from ab71a0c3d7 to ca84cb0e5d 2025-12-11 15:00:00 +01:00 Compare
nbelouin merged commit 48e83e1b1d into main 2025-12-12 08:19:46 +01:00
nbelouin deleted branch metal3-update 2025-12-12 08:19:46 +01:00
Sign in to join this conversation.
Reviewers
No Reviewers
steven.hardy
mchiappero
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
amorgante antaloala atanasdinov atrendafilov dbekhit dprodanov eminguez fdegirmenci geoagriogiannis ipetrov117 jarndt jtomasek kzhelyazkov mchiappero mkrutov nbelouin roxenham steven.hardy
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: suse-edge/Factory#317
Delete Branch "nbelouin/Factory:metal3-update"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?